Releases: auriocus/VecTcl
VecTcl 0.3 Checkpoint release
Public release with multimedia capabilities
This release of VecTcl includes a bridge to Tk photo images (see demonstration at EuroTcl) and a (limited) reader for sound files. A number of bugs was also fixed.
The binaries packages come in three flavours:
- vectcl-bin.zip contains just the VecTcl package with Tk extension for an unpatched Tcl
- VecTcLab contains a patched Tcl/VecTcl with a modified tkcon
- VecTcLab-bi additionally contains many useful packages:
-tcllib, tklib, Img, tklib_cur BWidget, gridplus, icons, memchan, mentry, nsf, TclRal, TclX, tdom, TkDnD, tkpath, tktable, tls (not on Windows), TkTreeCtrl, trofs (not on Win64), udp, wcb, XOTcl
The platforms are:
- Darwin: Mac OSX 10.5 - 10.10, 64bit
- Windows32: compiled using MinGW64 on Windows 10, tested on Windows 7
- Windows64: compiled using MinGW64 on Windows 10
- Linux32: compiled on Kubuntu 15.04
- Linux64: compiled on OpenSuSE 12.3
The binaries were compiled using kbs, and the modified kbs.tcl and package definitions are available in the build subdirectory of the master branch
Public test release
VecTcl public test release v0.1.7
This public test release demonstrates a large part of the functionality that VecTcl should provide. The supported vector language is close to the designed one, but the desired level of functionality is still not complete. Please expect bugs and performance issues.
This is an alpha release, so expect things to change without notice.
What already works
- Basic array shaping
- Most infix operators (elementwise and matrix product related)
- Integer, floating point and complex numbers
- Arbitrary rank tensors
- Reduction operators
- Some linear algebra math (FFT, SVD, equation solving, Eigendecomposition, Schur decomposition)
What needs to be done
- Implementation of a complete set of linear algebra math (matrix decompositions LU, Cholesky, pivoted QR, N-D FFT)
- Better syntax for array construction
- Performance improvements (operator folding, BLAS integration, compiler improvements, JIT compilation)
What was changed from 0.1.5
- lots of bug fixes and memory leaks
- Integration of LAPACK (no external dependency)
- \ works for singular matrices (thanks to LAPACK DGELSY)
- add more examples
Build issue fixed from 0.1.6
Note: version info in binaries is still 0.1.6, but functionalty is 0.1.7
Public test release
VecTcl public test release v0.1.6 DON'T USE
This public test release demonstrates a large part of the functionality that VecTcl should provide. The supported vector language is close to the designed one, but the desired level of functionality is still not complete. Please expect bugs and performance issues.
This is an alpha release, so expect things to change without notice.
What already works
- Basic array shaping
- Most infix operators (elementwise and matrix product related)
- Integer, floating point and complex numbers
- Arbitrary rank tensors
- Reduction operators
- Some linear algebra math (FFT, SVD, equation solving, Eigendecomposition, Schur decomposition)
What needs to be done
- Implementation of a complete set of linear algebra math (matrix decompositions LU, Cholesky, pivoted QR, N-D FFT)
- Better syntax for array construction
- Performance improvements (operator folding, BLAS integration, compiler improvements, JIT compilation)
What was changed from 0.1.5
- lots of bug fixes and memory leaks
- Integration of LAPACK (no external dependency)
- \ works for singular matrices (thanks to LAPACK DGELSY)
- add more examples
Public test release
VecTcl public test release v0.1.6
This public test release demonstrates a large part of the functionality that VecTcl should provide. The supported vector language is close to the designed one, but the desired level of functionality is still not complete. Please expect bugs and performance issues.
This is an alpha release, so expect things to change without notice.
What already works
- Basic array shaping
- Most infix operators (elementwise and matrix product related)
- Integer, floating point and complex numbers
- Arbitrary rank tensors
- Reduction operators
- Some linear algebra math (FFT, SVD, equation solving, Eigendecomposition, Schur decomposition)
What needs to be done
- Implementation of a complete set of linear algebra math (matrix decompositions LU, Cholesky, pivoted QR, N-D FFT)
- Better syntax for array construction
- Performance improvements (operator folding, BLAS integration, compiler improvements, JIT compilation)
What was changed from 0.1.5
- lots of bug fixes and memory leaks
- Integration of LAPACK (no external dependency)
- \ works for singular matrices (thanks to LAPACK DGELSY)
- add more examples
Public test release
Major speed-up in compiler
This release features much faster compilation (~20-30x vs. v0.1.0-alpha) due to a C based parser, as opposed to the Snit based parser from v0.1.0-alpha. This also removes all runtime dependencies. VecTcl now runs and builds on a bare Tcl8.6 installation. Only if the grammar is changed, tcllib is required.
For examples on usage and benchmarks please see the GitHub pages
Public test release
VecTcl public test release v0.1.0
This public test release demonstrates a large part of the functionality that VecTcl should provide. The supported vector language is close to the designed one, but the desired level of functionality is still not implemented in large parts, as well as performance improvements.
This release was compiled and tested under Linux 32bit, Linux 64bit, OSX 64bit (gcc and clang), Windows 32bit and Windows 64bit (mingw)
This is an alpha release, so expect things to change without notice.
What already works
- Basic array shaping
- Most infix operators (elementwise and matrix product related)
- Integer, floating point and complex numbers
- Arbitrary rank tensors
- Reduction operators
- Some linear algebra math (FFT, real SVD, QR, equation solving)
What needs to be done
- Implementation of a complete set of linear algebra math (matrix decompositions LU, Cholesky, pivoted QR, N-D FFT)
- LAPACK integration - probably as an infrastructure for the linear algebra
- Better syntax for array construction
- Performance improvements (via, e.g., operator folding, BLAS integration, compiler improvements, C based parser)